Patient response to a text messaging campaign to close colorectal cancer screening care gaps among primary care patients

Abstract Introduction Health systems are increasingly using digital health technologies to improve patient engagement and health outcomes. However, questions remain about who is served by this outreach and whether text outreach furthers the digital divide. This paper investigates a quality improvement study leveraging a text message-based intervention to increase colorectal cancer (CRC) screening rates among primary care patients. Methods We sent text messages to adults aged 45–75 who were identified in the electronic medical record as average risk for CRC, overdue for screening, and had visited a healthcare system primary care clinic in the last year. Patients were asked to verify if they were due for CRC screening. Those who were due for screening could select a preference for colonoscopy, Cologuard, or a call from a care coordinator. Our primary outcome was to describe the reach (total responded/total sent) of text messages. Our secondary outcomes were to assess differences in response patterns and explore the association between patient characteristics with preferred screening type. We used adjusted logistic regression models with backwards selection to identify associations between sociodemographics and patient responses. Results Among 11,108 patients who received text messages, 3,473 (31.3%) responded to the initial text with information about their CRC screening history. We found lower text responses among those who were older (OR = 0.99, 95% CI 0.98, 0.99) or were single (OR = 0.88, 95% CI 0.80, 0.98), and higher odds of response among those who were female (OR = 1.34, 95% CI 1.23, 1.46) or Black (OR = 1.14, 95% CI 1.04, 1.25). Among the 1,907 patients who identified they were due for screening, 1,319 requested a referral for a screening test or connection to a care coordinator: 35.9% requested direct referral to colonoscopy, 38.4% requested Cologuard, and 25.6% requested connection to a care coordinator. Conclusion Text messaging interventions can be an effective strategy to reach a broad primary care population about CRC screening, but reach was not equitable. Future work should explore how intentional strategies related to timing or language could mitigate engagement with text outreach, where care coordination is needed, and to assess impact on CRC screening completion.